Segmentation of Crack and Open Joint in Sewer Pipelines Based on CCTV Inspection Images

Abstract

Sewerage, one of major underground pipelines, is an important infrastructure for a modern city. In order to keep sewerage in a good structure and performance condition, planned routine inspection and rehabilitation are necessary. At present, image processing and artificial intelligence techniques have been used to develop diagnostic systems to assist engineers in interpreting sewer pipe defects on CCTV images to overcome human's fatigue and subjectivity, and time-consumption. Based on the segmented morphologies on images, the diagnostic systems were proposed to diagnose sewer pipe defects. This paper proposes a novel method of computer vision, morphological segmentation based on edge detection (MSED), to segment defects in sewer pipelines. In addition to MSED, the traditional image segmentation methods, including opening top-hat operation (OTHO) and closing bottom-hat operation (CBHO), were also applied to the defect segmentation. The historical inspection data revealed that crack and open joint were the two typical sewer pipeline defects in Taiwan, and the experimental result demonstrates that MSED and OTHO are useful for the segmentation of crack and open joint, respectively.